Open-Source DevOps · dograh-hq

dograh

Dograh is an open-source, self-hosted voice AI platform that lets you build and deploy conversational agents for inbound and outbound calling. It's positioned as an alternative to Vapi and Retell, with no vendor lock-in, flexible LLM/STT/TTS integration, and a visual workflow builder.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/dograh-hq/dograh

4.8k

GitHub stars

1.1k

Forks

Python

Primary language

BSD-2-Clause

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	dograh-hq/dograh
Owner	dograh-hq
Primary language	Python
License	BSD-2-Clause — OSI-approved
Stars	4.8k
Forks	1.1k
Open issues	21
Latest release	dograh-v1.41.0 (2026-07-06)
Last updated	2026-07-08
Source	https://github.com/dograh-hq/dograh

What dograh is

Python-based voice AI platform built on pipecat, supporting Speech-to-Speech, LLM-driven conversations, and telephony integrations (Twilio, Vonage, Telnyx). Containerized via Docker, MCP-native, with REST APIs and SDKs (Python, Node.js). Modular architecture enables custom model integration and source-level customization.

Quickstart

Get the dograh source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/dograh-hq/dograh.git2cd dograh3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Self-Hosted Inbound/Outbound Call Centers

Build lead qualification, customer support, or survey bots that run entirely on your infrastructure with zero data leakage to third parties. Control deployment location and data residency.

Custom Voice Agent Workflows with Flexible LLM Stack

Deploy agents using your own LLM (local or cloud), STT, and TTS services. Swap providers at the workflow level without vendor recertification or platform constraints.

Low-Latency Voice Processing in Regulated Industries

Financial services, healthcare, and legal firms requiring on-premises processing, audit trails, and full source code access for compliance audits and customization.

Implementation considerations

First startup takes 2–3 minutes to download container images; factor into initial deployment windows.
No API keys required for local testing (auto-generated), but integrating external LLM/TTS/STT requires provider credentials and configuration via dashboard.
Telephony integrations (Twilio, Vonage, Telnyx) are pre-built; adding new providers requires code contribution or custom integration work.
Test Mode and Web Call feature enable offline validation; verify agent behavior end-to-end before routing production calls.
QA Node available for prompt quality analysis; use during iteration to validate conversation design before launch.

When to avoid it — and what to weigh

Need Multi-Language Support Out-of-the-Box — Documentation explicitly states English support only; expanding to other languages requires custom work. Not suitable for global campaigns without development investment.
Require Managed SaaS with Guaranteed Uptime SLA — Self-hosting removes the managed guarantee. While a cloud version exists, proprietary SaaS players (Vapi, Retell) offer SLA-backed infrastructure and support at scale.
Prefer Minimal Operational Overhead — Self-hosting requires Docker, infrastructure management, monitoring, and updates. No turnkey solution; assumes engineering capacity for DevOps.
Need Mature Enterprise Support & Indemnification — Open-source project backed by founders, not a venture-scale SaaS org. No guaranteed SLA, commercial support tier, or legal indemnification stated.

License & commercial use

BSD 2-Clause (Simplified) License. A permissive, OSI-approved open-source license permitting use, modification, and redistribution in commercial and private contexts, provided original copyright and license notice are retained. No copyleft obligations.

BSD 2-Clause is commercially permissive. You may self-host and run Dograh in production for commercial purposes. However, no explicit commercial support, SLA, indemnification, or liability waiver is stated in the project. If commercial use is mission-critical, contact the maintainers (YC alumni founders) to clarify support tiers, liability terms, and any service-level commitments available.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Strong
Assessment confidence	High

Security considerations

Self-hosting data residency provides control over where voice/conversation data resides, eliminating third-party cloud exposure. No security audit, penetration test results, or formal security policy stated in the data. Anonymous telemetry is enabled by default (opt-out: ENABLE_TELEMETRY=false). Code review all integrations (especially external LLM/TTS/STT keys, telephony webhooks). Docker containers and dependencies should be scanned for vulnerabilities. No mention of encryption in transit/at rest, rate limiting, authentication strength, or compliance certifications (HIPAA, SOC2, etc.).

Alternatives to consider

Vapi

Proprietary SaaS, fully managed, enterprise support, guaranteed uptime. Trade-off: vendor lock-in, no source access, per-minute pricing, data on Vapi's cloud.

Retell

Proprietary SaaS with focus on conversational quality, managed infrastructure, per-minute billing. Trade-off: closed-source, no self-hosting, vendor lock-in, limited customization.

Twilio Flex + Twilio AI

Managed contact center platform with AI agent capabilities, strong telephony, enterprise SLA. Trade-off: higher complexity, vendor lock-in to Twilio ecosystem, SaaS pricing model.

Software development agency

Build on dograh with DEV.co software developers

Dograh gives you full control and transparency. Start with Docker in 60 seconds, customize your LLM/STT/TTS, and scale on your infrastructure. Explore the docs, try the cloud demo, or join the community Slack for guidance.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

canine sharkord gorilla

Related on DEV.co

Explore the category and the services that help you build with it.

Open-Source DevOps AI Application Development Cloud Deployment & DevOps API Development Talk to DEV.co

dograh FAQ

Can I use Dograh with my own LLM (e.g., Llama, OpenAI, Claude)?

Yes. Dograh is BYOK-enabled: you can bring your own LLM, STT, and TTS via the workflow builder. No lock-in to specific providers.

Does Dograh include telephony, or do I need to set up Twilio/Vonage separately?

Dograh includes pre-built connectors for Twilio, Vonage, Vobiz, and Cloudonix. You provide your own credentials for those services; Dograh routes calls through them.

What's the difference between self-hosted and the cloud version?

Self-hosted runs on your infrastructure (Docker); you manage ops and data residency. Cloud version (app.dograh.com) is managed by Dograh; requires cloud SaaS pricing model. Both use the same platform.

Is there commercial support or SLA?

Not explicitly stated in the project data. Open-source community support via Slack and GitHub. For commercial support, SLA, or indemnification, contact the founders directly.

Software developers & web developers for hire

From first prototype to production, DEV.co delivers software development services around tools like dograh. Our software development agency staffs experienced software developers and web developers for custom software development, web development, integrations, and ongoing support across open-source devops and beyond.

Ready to Deploy a Self-Hosted Voice Agent?

Talk to DEV.co Schedule a Consultation